-
Notifications
You must be signed in to change notification settings - Fork 46
/
metadata-UVA-adaptive.txt
26 lines (26 loc) · 1.99 KB
/
metadata-UVA-adaptive.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
team_name: UVA Biocomplexity Institute
model_name: adaptive
model_abbr: UVA-adaptive
model_version: "v23"
model_contributors: Przemyslaw Porebski (UVA) <[email protected]>, Srinivasan Venkatramanan (UVA) <[email protected]>, Bryan Lewis (UVA) <[email protected]>, Aniruddha Adiga (UVA) <[email protected]>, Brian Klahn (UVA), Benjamin Hurt (UVA), Jiangzhuo Chen (UVA), Anil Vullikanti (UVA), Madhav Marathe (UVA)
website_url: https://biocomplexity.virginia.edu/project/covid-19-pandemic-response
license: cc-by-4.0
methods: We use a discrete time SEIR compartmental model with explicit strain modeling fit to observed hospitalizations at the state level. The fitting procedure takes into account variable hospitalization rate, depending on the immune status derived from round 16 results.
Deaths are projected using rescaling of the projected hospitalizations by the current (last 3 weeks) death-to-hospitalization ratio per state. Uncertainty bounds are obtained by using an experiment design over other model parameters such as infectious/incubation periods, reporting delay and bounds on hospitalization rates. Representative trajectories are selected using K-medoids clustering.
modeling_NPI: NPI impact and adjustments are done by proxy using the fitted transmissibility. The current level of transmissibility after modeling strain effects is assumed in the future.
compliance_NPI: "Not applicable"
contact_tracing: "Not applicable"
testing: "Not applicable"
vaccine_efficacy_transmission: "Not applicable"
vaccine_efficacy_delay: "Not applicable"
vaccine_hesitancy: "Not applicable"
vaccine_immunity_duration: "4 months"
natural_immunity_duration: "4 months"
case_fatality_rate: "Not applicable"
infection_fatality_rate: "Not applicable"
asymptomatics: "Not applicable"
age_groups: "Not applicable"
importations: "Not applicable"
confidence_interval_method: "Not applicable"
calibration: "Fitting weekly transmissibility to match weekly hospitalizations "
spatial_structure: "State-level, not transmission between states"